New technologies and data science are transforming a wide range of organizations into analytics-intensive enterprises. Despite the resulting demand for graduates with experience in the application of analytics, though, undergraduate education has been slow to change. The academic and policy communities have engaged in a decade-long conversation about the need for and more effective ways to boost the number of college graduates with degrees in science, technology, engineering, and mathematics (the STEM fields). But this broader transformation will require many more college graduates, including those with non-STEM degrees, to be proficient in the application of analytics in their fields. This need provides higher education with an opportunity to take a fresh look at the skills of liberal arts graduates and the value of liberal arts degrees. This article highlights the following points: (1) Employers from all sectors seek "data-enabled" professionals who possess a deep understanding of--and are current on analytical methods relevant to--their disciplines; are familiar with technology; and can interact, analyze, communicate, and work with people from other disciplines and cultures; (2) By confining data science to graduate and upper-level undergraduate courses in engineering or computer science, colleges and universities are inadvertently reducing the diversity of the pool of data-science-enabled professionals; (3) Liberal-arts programs need to include the development of STEM skills in non-STEM graduates. Some institutions have accordingly created majors, minors, concentrations, and digital-infused curricula in disciplines such as library science, media, and the arts; and (4) Graduate programs aimed at individuals with data-science expertise gained through work experience and with skills in both STEM and non-STEM fields are rapidly expanding, particularly in business and health care. The Business-Higher Education Forum has helped develop both graduate and undergraduate programs in cybersecurity that can serve as models.